Abstract
Abstract : The main objective of this dissertation is to identify the key determinants of industrialisation in Ethiopia over the period 1981 to 2014. Industrialisation is measured as the share of manufacturing value added (MVA) in GDP. Ten potential determinants of the share of MVA are considered within the framework of a general, unrestricted auto-regressive distributed lag (ARDL) model. To select the parsimonious model, the dissertation uses a general-to-specific (Gets) model-selection algorithm based on Hendry’s (1995) methodology. The study finds the following significant results. First, there is an inverted U-curve relationship between GDP per capita and the share of MVA in Ethiopia. This is in line with the theory, in which demand moves away from manufacturing to service activities over time, and/or technological advancement that reduces the relative price of manufacturing goods. Second, the size of the domestic market positively affects MVA’s share in Ethiopia, but beyond a certain threshold level it has a negative effect. Third, in contrast to conventional wisdom, an appreciation of the real exchange rate (𝑅𝐸𝑅) stimulates Ethiopia’s manufacturing sector. This may be attributed to Ethiopia’s large domestic market and the importation of cheap capital and intermediate goods that are essential inputs in the manufacturing sector. This strategy, however, relies heavily on the ability of the economy to attract long-term capital inflows other than foreign borrowing to finance the resulting import-export gap. It follows that, once the manufacturing sector is mature enough to compete in world markets, policymakers need to maintain the 𝑅𝐸𝑅 at competitive levels. Fourth, foreign direct investment (FDI) has a positive and significant effect on Ethiopia’s manufacturing sector. Thus, Ethiopia should design appropriate policy measures to attract more FDI.
M.Phil. (Industrial Policy)